M6-040 850 20 01 02

ltr
a
W
s
n
m
ie
S
MAGNETOM
MR
Function Description
System
Tune-up / QA
Siemens, 2008
10276755
Siemens, 2008
- For internal use only - All documents may only
be used by authorized personnel for rendering
services on Siemens Healthcare Products. Any
document in electronic form may be printed
once. Copy and distribution of electronic docu-
ments and hardcopies is prohibited. Offenders
will be liable for damages. All other rights are re-
served.
Print No.: M6-040.850.20.01.02 English

Replaces: n.a. Doc. Gen. Date: 03.08
CS TC MR
n.a.
2008
2 Copyright / Version / Disclaimer
1Copyright / Version / Disclaimer
Copyright
Siemens, 2008 refers to the copyright of a Siemens entity such as Siemens Aktienge-
sellschaft - Germany, Siemens Mindit Magnetic Resonance Ltd. - China, Siemens Shang-
hai Medical Equipment Ltd. - China, Siemens Medical Solutions USA Inc. - USA and/or
Siemens Healthcare Diagnostics Inc. - USA.
Document Version
Siemens reserves the right to change its products and services at any time.
In addition, manuals are subject to change without notice. The hardcopy documents corre-
spond to the version at the time of system delivery and/or printout. Versions to hardcopy
documentation are not automatically distributed. Please contact your local Siemens office
to order current version or refer to our website http://www.healthcare.siemens.com.
Disclaimer
Siemens provides this documentation as is without the assumption of any liability under
any theory of law.
The service of equipment described herein is to be performed by qualified personnel who
are employed by Siemens or one of its affiliates or who are otherwise authorized by Sie-
mens or one of its affiliates to provide such services.
Assemblers and other persons who are not employed by or otherwise directly affiliated with
or authorized by Siemens or one of its affiliates are not entitled to use this documentation
without prior written authority.
MAGNETOM M6-040.850.20.01.02 Page 2 of 88 Siemens, 2008

03.08 CS TC MR For internal use only
Table of Contents 3
0Table of Contents
1 _______ General ________________________________________________________ 6
The Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Procedure Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Reports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Quality Assurance Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Active Coil Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2 _______ RF-Related Tune-Up_____________________________________________ 14
Tuning Calibration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Body Coil Tuning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Adjust Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
RF Characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Effects of amplitude distortions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Effects of phase distortions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Method of correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
RF Characteristic LC (Local Coils) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Coil Power Losses (CPL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3 _______ Gradient-Related Tune-Up________________________________________ 30
Regulator Adjust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Table Adjustment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Phantom Shim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Siemens, 2008 M6-040.850.20.01.02 Page 3 of 88 MAGNETOM

For internal use only 03.08 CS TC MR
4 Table of Contents
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Cross Term Compensation (CTC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Eddy Current Compensation (ECC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Spatial Dependency of Eddy Currents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Gradient Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Gradient Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Nominal Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Actual Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Table Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4 _______ Receive Path-Related Tune-Up ____________________________________ 56
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Relative Receive Path Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Absolute Receive Path Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5 _______ General Quality Assurance _______________________________________ 63
Image Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Gradient Rise Time Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Calculation Artefacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Spike Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Physical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Stability Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Table of Contents 5
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Fat Saturation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Evaluation of Measured Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
SAR Monitor Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Synthesizer Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Principle of Synthesizer Test. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Temperature Sensor Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Expert Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Stability_LongTerm Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Evaluation of Measured Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Field Stability Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Principle and evaluation of MR-measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Evaluation of Field Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
6 _______ Coil Quality Assurance __________________________________________ 80
7 _______ Appendix _____________________________________________________ 81
Shim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
The model: Spherical Harmonics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Some examples for inhomogeneities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Array Shim Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
8 _______ Changes to previous version _____________________________________ 87

6 General
1-
1General
The Tune-Up is a collection of procedures designed to calibrate, compensate or to deter-

mine hardware variables which need post-correction. They are performed initially to tune
the MR system after a new installation. Some Tune-Up procedures will need to be per-
formed after the repair and/or replacement of certain components as described in the
Replacement of Parts procedures.
This description explains the physical background of the individual Tune-Up procedures.
The more simplified and automatized the service-software has become, the more impor-
tant it is to have a good understanding of these procedures and what they are doing.

General 7
The Procedures 1.1
The procedures fall into three major parts:

RF-related Tune-Up Procedures
- Tuning Calibration
- Body Coil Tuning
- RF Characteristic
- RF Characteristic LC (Local Coils)
- Coil Power Losses (CPL)
Gradient-related Tune-Up Procedures
- Regulator Adjust
- Table Adjustment
- Phantom Shim
- Cross Term Compensation (CTC)
- Eddy Current Compensation (ECC)
- Gradient Delay
- Gradient Sensitivity
Receive Path-related Tune-Up Procedures
- Relative Receive Path Calibration
- Absolute Receive Path Calibration
General Quality Assurance
- These measurements mainly verify the previously adjusted Tune-Up parameters.
- In addition, a set of general image quality- and system stability-evaluations are per-
formed, mainly using the Body Coil.
Coil Quality Assurance
- These procedures check Signal to Noise and Image Brightness for the individual
coils with phantom measurements.

8 General
Tune-Up and QA Procedure Steps
Fig. 1:
General 9
Dependencies 0
All selected procedures are performed from top to bottom. The individual steps follow a
specific order and contain dependencies between those prior and those that follow. The
order could be changed, but the dependent steps will get the state "ToDo", which might
lead to perform some steps again; so it's a good idea to stay with the order.
If a Tune-Up procedure has changed its status, also the relevant procedure(s) in the Quality
Assurance menu for the corresponding coil are changed to To Do.

10 General
Modes 1.2
Each of the Tune-Up steps can be performed in two different modes:

Normal mode: this is the mode that has to be used to achieve a status of "Done".
Expert Mode: This mode is intended for troubleshooting only. Procedures which can
be run in this mode offer the possibility to change or make individual selections of
sub-steps of a procedure. For example, a single gradient axis could be measured alone
or certain parameters can be selected. The determined or measured values will also be
saved in this mode!
NOTE The Tune-Up procedures in Normal mode measure and auto-

matically SAVE system parameters, so they should NOT be
used for trouble shooting or if the system is not in proper
working order. In that case, the Quality Assurance proce-
dures should be used to VERIFY system specifications with-
out saving any system parameter.

General 11
Procedure Status 1.3
In a fresh installation all procedures will be set to To Do.

If a procedure was done successfully and all values are in specification it will get the Done
status.
In case a procedure has stopped or aborted due to an error or time-out, it will be set to
Error.

12 General
Reports 1.4
For every procedure, the results are stored as an HTML file and can be called up for display
under "Reports", "Session History". The saving of the measured data in the system param-
eter files, log files and data files will be done automatically if the procedure was performed
successfully.

General 13
Quality Assurance Procedures 1.5
The QA procedures are a set of procedures which are used to verify system performance
and include all the procedures found in the Tune-Up as well as additional procedures for
testing overall system performance. Only these will be described since those procedures
found in Tune-Up are identical, with the exception that no values are saved:
If you are out of specification after a Quality Assurance step, you must perform the cor-
responding Tune-Up step.
Every Tune-Up and Quality Assurance procedure can be run in either Normal- or
Expert-Mode.
Active Coil Menu 0
The Active Coil menu shows the name of the currently selected coil that will be used by the
Tune-Up procedures. Since the body coil is always connected, it will always be the default
selection when the Tune-Up platform is opened. In addition to the Body Coil, any number
of local coils may be connected to the system at any one time. These coils are displayed
in the Active coil pull-down menu. The first coil displayed in the list is always the Body Coil.
If no coil is connected to a plug the message "No Coil" replaces the coil name.

14 RF-Related Tune-Up
2-
Tuning Calibration
2RF-Related Tune-Up
Fig. 2: Normalization of Tune Circuit

The circuit used to test the Body Coil must first be calibrated before it can be used. The
calibration is normalizing the impedance characteristics of the components in the Body Coil
Measurement Circuit which includes the TX_Module, BCCS, TALES, RCCS, RX_Module
and all connecting cables.

RF-Related Tune-Up 15
Measurement 0
The TALES output will be terminated with an open, a short and a 50 ohm termination which
covers the complete impedance resistive range. In a last measurement step the
BC-system 1 and BC-system 2 output of the TALES has to be connected together by
means of the standard 2 m BNC service-cable of known impedance.
Under these different impedance conditions the forward and reflected values are measured
by applying a rectangular RF pulse (tuncal sequence) to the Ur and Uf side of the direc-
tional couplers in the BCCS. For each termination condition, the frequency is varied
through the complete operating frequency bandwidth in 75 steps.
Since there is also a coupling of the two Body Coil systems over the Rx Hybrid which has
an effect on the impedance, it must be measured and compensated as well. This is done
by measuring the BC-system 2 path when sending with the BC-system 1 and vice versa.
The coupling factors are expressed as parameters of the H-matrix..
After this measurement, the Tune Measurement Circuit will be normalized - the point of
measurement (POM) (i.e., the point where the system knows the exact amplitude, fre-
quency and phase of the signal being sent or received) will be transferred to the Point of
Interest (POI) and the system is now able to "see" the impedance conditions at the POI.
Expert Mode 0
In Expert Mode single measurement steps can be selected and measured.

Results 0
Fig. 3: Results of Normalizations

A graphic displays the results of the parameters H11, H12, H21, H22 (Rx hybrid coupling).
Each colored line represents one of the four measured H parameters with each line dis-
playing the 75 measured frequencies (tick marks). It is difficult to interpret this graphic but
it may help if compared with a measured which was acquired when the system was func-
tioning properly.
The tuning calibration is also verified. This is accomplished with two measurements. The
first measures a 50 ohm termination at the BC-system 1. If the tuning calibration measure-
ment was correct, both real and imaginary parts for 50 ohms should be zero (= 0 % reflec-
tion). The second verification measures a short (0 ohms) at the BC-system 2 which should
give real = -1, imaginary = 0 (100 % reflection).

Body Coil Tuning 2.1
This procedure is used to measure and, if necessary, adjust the central resonant frequency
of each of the two Body Coil systems as well as minimize the coupling between them.
Measurement 0
The measurements are performed over an 800 kHz frequency range. No load is used for
this measurement, therefore the table must be retracted out of the bore. The resonant fre-
quency is found at that point where the reflection of the BC has a minimum. The decoupling
is given by the frequency where the transmission has its maximum value. This should be
lower than -18 dB (-18 dB = 1.6 % power loss) to minimize losses.
Evaluation 0
Resonance Frequency
The procedure measures the frequency response of the Body Coil over an 800 kHz fre-
quency range.
The evaluation determines:
if the center frequency of each Body Coil system is between 122.95-123.45 MHz
if the center frequencies of both coil systems are tuned within 100 kHz delta
that the reflection factor is less than 65% (r 0.65)
Transmission (coupling)
The transmission through the Body Coil is the amount of coupling between the two Body
Coil systems due to coil construction. It is measured by sending a signal into the
BC-system 1 and measuring what comes out of the BC-system 2.
Adjust Procedure 0
This is a manual adjustment requiring the front tunnel cover to be removed. On the Body
Coil are three 60-turn trim capacitors (refer to next Figure), two for adjusting the frequency
of each Body Coil system and one for minimizing the coupling between the two.
Tuning
The trim capacitors T1 and T2 vary the resonance frequency of the system 1 and 2 respec-
tively. Turning the cap clock-wise LOWERS the frequency, counter clock-wise INCREASES
the frequency.
Decoupling
TD is also a 60-turn capacitor and effects the de-coupling between the two BC systems. It
is not possible to know which direction to take at first, but there is only one minima.

NOTE There is a interdependency between all three adjustments!

Refer to the SeSo Online Help for a detailed description of
this procedure
Expert Mode 0
There is no Expert Mode for this procedure.

Body Coil Resonance Frequency and Decoupling Adjustments
Fig. 4:
RF Characteristic 2.2
Fig. 5: Pre-distortion of RF-pulses

The calibration of the RF transmitting system is important for obtaining optimum image
quality. The RF pulse amplitude defines not only the flip angle, but the pulse-lobe amplitude
relationship also defines RF pulse profile and hence the quality of the slice profile.

Effects of amplitude distortions 0
Amplitude distortions of the RF envelopes caused by the nonlinearity of the amplifier may
produce distortions of the slice profile and slice thickness with the following consequences:
reduced signal to noise caused by crosstalk (in multi-slice sequences)
reduced contrast due to the resulting partial volume effect
The principle of amplitude corrections is shown in the previous graphics.
Effects of phase distortions 0
A phase distortion would lead to a shift or drift in the frequency, which would result in false
slice positions, hence the phase distortion must also be corrected.
Method of correction 0
The distortion caused by the non-linearity of the RF amplifier is corrected by measuring the
output of the RF transmission over the known input to the RF amplifier for the entire oper-
ating range of the RF amplifier. A corresponding inverse function of the amplitude and
phase characteristics is used for pre-distortion of the input signal in order to achieve a lin-
ear output characteristic and to optimize the RF excitation.

Measurement 0
Fig. 6: RF Characteristic Measurement

The RF-characteristic-measurement is performed with the following steps:
1. Calibration Voltage Search: Starting with a nominal value of 100 V, the output is
increased until a value of 750 V is measured at the TALES.

2. RF Calibration: The characteristic measurement will be made using the DICO, so it

must first be calibrated. Using the results of the first step, the DICO value will also be
measured giving a relation between the TALES and DICO. This value is used as a cor-
rection factor for the DICO values.
3. RF Max Power Search: The system now increases the input voltage to the RFPA until
an output power of 28.322 kW is measured at the TALES.
4. RF Characteristic measurement: For the following measurement the RF is sent to the
50 ohm dummy in the TAS. 256 pulses with different amplitudes are applied. The first
pulse starts with the maximum voltage and is decreased by 0.25 dB for each subse-
quent pulse. This ensures that the complete characteristic is measured.
5. Save the RF Characteristic result in the rfchara.dat file.
6. Verify: The final step is to perform a verification measurement. The inverse of the mea-
sured characteristic curve is now used to pre-distort the RF pulses. The result should
be a linear output transfer characteristic of the RFPA.
7. TAS-Dummy conversion factor: Additionally the TAS-Dummy conversion factor,
which characterizes the attenuation of the path DICO-TALES-TAS, is determined and
saved. TAS-Dummy conversion factor = Voltage Dummy / Amplitude Dummy with:
- Voltage Dummy = 10**(TAS-TALES_TAS-Damping/20 dB) x Amplitude DICO *
DICO factor
- Amplitude DICO = Voltage, obtained from DICO forward during TAS dummy calibra-
tion measurement
- Amplitude Dummy = ADC value, obtained from PD signal during TAS dummy cali-
bration measurement.

Results 0
Fig. 7: RF Chara Results
Plot of Measured Characteristics

The first set of graphics show the measured amplitude and phase characteristic curves,
each measured with 256 measure points and normalized RF input.

Plot of Calculated Characteristics

The second set of graphics displays a calculated version of the measured data which is
necessary for the RF Characteristic data file.
Plot of Verification Measurement

The last set of graphics displays the evaluation results of the verification measurement. The
verification measurement is made by using the inverted characteristic curve to pre-distort
the RF pulses. For a correct compensation, the result should be a linear input to output
transfer characteristic. The evaluation is made by calculating the deviation of the measured
data to an ideal linear function which is derived by simply creating a straight line between
the first data point (0V) and the last data point (maximum voltage). A check for monotony
is also made, assuring that the amplifier at no time went into saturation.
Expert Mode 0
The Expert Mode allows the user to modify the maximum voltage.

RF Characteristic LC (Local Coils) 2.3
Non-spectroscopy systems:
- For measurement of the Local Coil characteristic the Service Plug must be con-
nected to the coil plug 1.
- In addition, a manual cable connection from the LC TX-output of the TALES to the
dummy load in the TAS must be established.
Spectroscopy systems:
- In addition to non-spectroscopy systems all RF-amplifiers and available nuclei (H
and all X-nuclei) in the Local coil (LC) path will be measured.
Measurement 0
The dummy load in the TAS_C must be connected to the LC-output of the TALES (i.e. dis-
connect output cable from TALES X6, disconnect input cable at BCCS X7 and connect it
with TALES output X6) and the Service Plug must be connected to patient table coil plug 1.
The RF characteristic is measured as follows:
1. Calibration Voltage Search: Starting with 100 Volts output (this is just a nominal
value). The TALES readings are used to increase the output until 400 Volts output are
measured.
2. RF Calibration: The DICO has to be calibrated since it is used for the final measure-
ment of the characteristic. The DICO value is measured, based on the resulting input
voltage from step 1. The DICO is calibrated using the known relationship between
TALES and DICO.
3. RF Max Power Search: By carefully increasing the output voltage, the max. power of
3581 W, respectively 6685 W (depending on nucleus) is searched.
4. RF Characteristic measurement: For the following measurement, the RF is sent to
the 50 ohm dummy in the TAS. 256 pulses with different amplitudes are applied. The
first pulse starts with the maximum voltage, decreasing by 0.25 dB for each subse-
quent pulse. This ensures that the complete characteristic is measured.
5. The RF Characteristic is saved in the rfchara.dat file.
6. The final step is the verify measurement. This includes the calculation of amplitude
and phase deviation and checking if the results are within specification. The results and
graphics can be viewed in the service software under Tune-Up, Report, Tune-Up
Results.
Results 0
Same as for RF Characteristic procedure for Body Coil.
Expert Mode 0
In Expert Mode, the function operates similar to standard mode except that the following
options are available:
1. You can click Save to save the RF Characteristic after measurement.

2. You have an option to modify the "Maximum allowed RF power".

3. Selection of RF-amplifier, coil path and nucleus.
4. Option to switch off the frequency adjustment
5. Option to disable switching of TAS to dummy load.
Also, Expert Mode does not include RF-verification at the end of the measurement.

Coil Power Losses (CPL) 2.4
Fig. 8: Coil Power Losses Measurement

The purpose of this procedure is to measure the amount of power loss in the Body Coil.
This value is used by the RF-Safety Watchdog (RFSWD) monitor to achieve a more accu-
rate SAR measurement.

Measurement 0
A complete inline adjustment with a non-volume-selective frequency adjustment is per-

formed, followed by a transmitter adjustment. The voltage from the Adjust/Transmitter
result is used to calculate the reference power:
Reference Power from TALES = (Voltage from Adj/Tra)2 / 50 .
The small 170 mm spherical phantom is necessary because of the reduced losses in the
phantom and the dielectric resonance effects.
A subsequent measurement is made only to verify that the correct phantom type is used
and that is it centered.
The following data are acquired:
Pref TALES from Adjust Transmitter
BC Pickup probe 0 value
BC Pickup probe 1 value
Evaluation 0
The measured reference power from the TALES is checked. Low Spec = 500 W, High Spec
= 2600 W. If a wrong phantom were to be used the low specification might not be reached
or if the body loader phantom is in the magnet the high specification might be exceeded.
Results 0
The value for the Coil Power Losses is calculated: CPL = Reference Power from TALES *
(1 + ), where is a correction factor taking into account electrical conductivity of the phan-
tom fluid and dielectric resonance effects. The factor is greater than 1 so the Coil power
loss value is larger than the measured reference power.

30 Gradient-Related Tune-Up
3-
Regulator Adjust
3Gradient-Related Tune-Up
Fig. 9: Regulator Adjustment

The regulators in the GPA have a proportional, P, (= gain), integral, I, (= time) and differ-
ential, D, regulation characteristics, thus each regulator consists of a P, I and D regulator
circuit in series. The purpose of this procedure is to find the optimal P, I and D regulator
adjustment values for an optimized regulation characteristic.

Gradient-Related Tune-Up 31
Measurement 0
The regulators are tuned one after the other in the order PID. Each regulator begins with a
configured start value or in Expert Mode with the user given start value. The regulator value
then is incriminated by one single bit and a measurement is done. The measurement
results are evaluated and a new regulator value is calculated. This step is repeated until
the regulator is optimized. Finally, a configured safety factor is added to the regulator.
The actual value gradient waveform is sent over the gradient loop path from the D60 CAN
& Service board in the GPA to the PCI_MON board in the AMC for evaluation. The Gra-
dReg program then performs the following:
checks of measured data
displays the measured pulse: the first graphic shows the complete gradient pulse, the
second graphic displays the overshoot of the gradient pulse
performs an evaluation (described below)

Evaluation 0
Fig. 10: Regulator Siignal Evaluation

The pulse is evaluated for three criteria:
Overshoot amplitude to the nominal pulse amplitude (in %)
Decay time for overshoot to decay to 15% of its amplitude

Smoothness is the maximum positive deviation of the gradient pulse top portion.
The smoothness of the overshoot is determined in this way: Starting from the maximum
data point of the overshoot, the difference to the next successive data point is determined.
This is repeated for all data points of the overshoot. The smoothness factor is the maximum
difference determined by the above process divided by the nominal pulse amplitude and
expressed in %.
Pulse Check
A check if the regulator is being over-driven or the pulse signal oscillates is also made, indi-
cated by 2 or more sign changes in the data points of the overshoot. If there are points with
an amplitude below the nominal amplitude, an undershoot after an existing overshoot is
detected. That means over-driven regulator or oscillating signal.
Results 0
The final regulator adjustment values and achieved values are displayed in tabular form.
Fig. 11: Regulator Adjust Final Results

After all regulators (PID) are tuned, it is checked if the signal is in specification. If so, the
regulator values are set permanently. If an error occurs or the program is aborted the old
regulator values (saved before Tune-Up starts) are restored for the GPA.
Expert Mode 0
If the factory default values are too tight and cause the regulator to oscillate and result in
an aborted measurement, the default values can be changed in the Expert Mode menu.

Table Adjustment 3.1
Fig. 12: Table Adjustment Results

The goal of the Table Adjustment is to determine:
center positioning of the phantom in the magnet iso-center
sufficient field homogeneity

distance between light marker and magnet iso-center

The Phantom Shim, CTC and ECC procedures require a centered phantom for accurate
results. The Table Adjustment determines the phantom position with respect to the center
of the Gradient Coils (iso-center). From this, the distance from the light marker to the mag-
netic iso-center is determined.
Measurement 0
The measurement uses a double echo sequence. The first echo is acquired without apply-
ing a readout gradient resulting in an FID the length of which is a rough check of the fun-
damental imaging volume homogeneity.
During the second echo, a readout gradient is applied resulting in a signal spectrum of the
phantom. The center of the phantoms signal spectrum is compared to the center of the
scan volume frequency window to calculate the phantom position in readout direction. The
sequence measures all three axes in this way however the previous figure shows as exam-
ple only the X axis.
Results 0
The results of the Z axis measurement is used to determine the table position and will be
corrected automatically and stored as the new light marker distance. All three phantom
position values are displayed in a table. A necessary correction in the X direction can be
made by re-positioning the phantom. Corrections in the Y axis may be made by adjusting
the table height.

Phantom Shim 3.2
Fig. 13: The 3D - Shim Sequence

Phantom Shim measures magnet field inhomogeneities and calculates the gradient offset
and shim coil currents (option) necessary for optimization. It is performed at this point in
the Tune-Up as preparation for the following procedures using the ball phantom. The cor-
rection currents determined here also serve as default values for the 3D-Shim procedure
used for patient imaging.

Background 0
The phase of an MR signal is directly proportional to the magnetic field. Phantom Shim
makes a 3D-measurement, a so-called "Field Map" using a 32 x 32 x 32 voxel matrix cov-
ering a 35 x 35 x 35 cm volume.
An advantage of this method is to acquire field values for each of the voxels which later
allows the user to define any specific local volume optimization during 3D-Shim adjust-
ments with patients. To reduce calculation times, the evaluation algorithm detects voxels
with little or no signal which are simply discarded.
In order to reduce hardware effects (e.g., gradient delays or multiple coil element phase
errors) two echoes are measured in one 3D-Dess-sequence (Double Echo Steady State)
and the phase difference of both echoes is analyzed. Then, the measurement is repeated
a second time with opposed gradient polarities to eliminate eddy current effects.
Measurement 0
Shim System Check

First, an operational check of the shim hardware is made. A nominal shim current is applied
to each shim channel followed by a shim measurement and the actual field change effect
is compared to the expected, predicted field change. In this way wrong cabling, wrong
polarity, open or short coils can be detected.
Field Map
The sequence is a 3D-DESS-sequence, i.e. two echoes, a FISP and a PSIF echo are gen-
erated. The echo times - 4.7 ms - are chosen for fat and water in phase, suitable for patient
use as well. As you can see in the next figure, a static field inhomogeneity will result in a
phase rotation of the magnetization vectors - this is how the shim situation can be deter-
mined.
Evaluation 0
For the check and optimization of the B0-field it is sufficient to compare the field differences
between neighboring voxels. The field difference is obtained by comparing the phases of
the magnetization vectors. A shift of 10 degrees from voxel to voxel in one direction results
in a linear homogeneity of about 9 T/m in that direction.
The complete data set is evaluated by a differential shim equation.The equation minimizes
the difference between:
the field generated by the three Gradient Coils and the 5 shimcoils
the measured magnetic field inhomogeneities.

Fig. 14: Phase rotation of magnetization vectors in the voxels

Numerically, the problem is a least square fit leading to a system of many linear equations
with 8 unknowns: the 3 gradient offset currents and the 5 shim currents. In short: the com-
parison of the B0-fields in two voxels - performed by analyzing the phases - results in one
equation. For all voxels 98304 calculations are required. However, a threshold is set for the
signal to noise: voxels with no or little signal-noise ratio will be rejected, so the number of
equations is typically reduced to about 20000.
Results 0
The first table displays the measured field terms for each of the 8 terms that can be cor-
rected. The 3 first order field terms and the 5 second order field terms that can be optimized
are:

Tab. 1 Spherical Harmonics
Spherical Field Correction

Harmonics Device
First Order Terms
A(1,0) Gz Z Gradient Coil
A(1,1) Gx X Gradient Coil
B(1,1) Gy Y Gradient Coil
Second Order Terms
A(2,0) z2 - (x2+y2)/2 Shim Coil
A(2,1) xz Shim Coil
B(2,1) yz Shim Coil
A(2,2) (x2-y2)/2 Shim Coil
B(2,2) xy Shim Coil
The following two tables give the correction currents for the gradient and shim coils. The
gradient offset current are expressed in mT/m and the shim coil currents in mA.

Fig. 15: Phantom Shim Results
Expert Mode 0
No extra functionality is provided by the Expert Mode.

Cross Term Compensation (CTC) 3.3
Fig. 16: Cross Term Currents

An applied gradient field in any axis produces not only eddy currents along that axis, but
also small amounts of eddy currents which are coupled over the conductive magnet com-
ponents onto the other two axes, the so-called cross terms. The CTC (cross term compen-
sation) procedure measures the eddy currents created by the cross terms. Compensation

is made by pre-distortion of the gradient waveform. Unlike the main eddy currents which
are complicated and need to be modeled with 5 time constants, the Cross Term current
decay exponentially in time, generally with one predominant time constant only.
Measurement 0
The procedure measures the decay of the eddy currents caused by the cross terms. Ampli-
tudes and time constants of the eddy current contributions are then calculated. The com-
pensation is straightforward: suitable correction gradient fields are applied perpendicular to
the main applied gradient field during imaging sequences. The result is a much cleaner
main gradient with very small residual cross terms improving the final image quality.
Result 0
The amplitude and time constants if the compensations are displayed graphically and in
tabular form. In the previous figure only the Z axis is shown.
Expert Mode 0
The CTC can be made individually for each axis.

Eddy Current Compensation (ECC) 3.4
Fig. 17: Eddy Current Measurement

The dynamic gradient fields produce currents (called eddy currents) in all the surrounding
conductive structures, mainly the Body Coil, magnet bore and the cryo-shields. The eddy
currents produce in turn magnetic fields which oppose and distort the applied gradient

fields. Eddy currents in the warm components (Body Coil and Magnet bore) have relatively
short decay times, whereby eddy currents developed in the cold cryo-shields can have rel-
atively long time constants.
The eddy currents have to be accurately quantified and characterized by their time con-
stants and amplitudes to achieve a good compensation. The compensation is made by
adding the reciprocal of the measured eddy currents to the gradient pulse thus neutralizing
their effects. For accurate compensation 5 time constants are used.
Spatial Dependency of Eddy Currents 0
Experience has shown, that for smaller slice shifts it is sufficient to express the eddy fields
into a 0th and a 1st order term.
Tab. 2 Eddy Current Terms
Term Description
This term arises from an asymmetry of the Gradient Coil with respect to the
magnet bore and cryo shields. This term is space independent and present
0th in the complete imaging volume and adds to the nominal B0-Field. The ampli-
order tude (x) is given in the unit Tm/mT. The time constant of the most significant
0th term component is about 500 ms, although there are shorter time con-
stant components as well.
1st This is the most important term as it has the same symmetry as the gradient
order field itself. The amplitude is given in % of the applied gradient pulse.
2nd order There are also high ordered eddy currents present, they are usually small
and negligible as long as the slice shifts and off center zooms are not too
or above large. No compensation for the high order eddy currents can be made.

Measurement 0
Fig. 18: ECC Preparation Measurement

First, the phantom center position is checked. The same preparation measurement is used
as for Phantom Shim, however, the small spherical phantom is used. If the phantom is not
correctly positioned, the measurement stops with an error.
The sequence for the measurements generates 23 Spin Echoes with delay times between
0.4 ms to 9000 ms of the applied gradient.
The measurement is made at the moderate slice shifts of 5 cm to prevent contributions
from higher order terms.

Fig. 19: Eddy Curremt Measurement

Evaluation 0
Fig. 20: ECC Evaluation

After a measurement - usually two or three iterations are required before the final optimi-
zation - the measured data is displayed in a table found in the log file. The first table (not
shown) lists the actual measured data at position +50 mm from the first measurement. The
second table lists the measured data for the -50 mm position. The third table, shown above,
gives the combined results of both.

Header
There is some basic information given in the header (top section). For example, the
sequence generated 23 Spin Echoes. The delay times can be seen in the list and are
between 0.4 ms to 9000 ms of the applied gradient. The gradient amplitude is 8 mT/m, the
Spin Echoes are generated in 3 mm slices with slice shifts of +50 mm and -50 mm. Hence
the gradient amplitude in the slices is:
8 mT/m 0.05 m = 0.4 mT
corresponding to a frequency of
0.4 10-3 42.577 106 Hz = 17031 Hz
at the slice position +50 mm or - 50 mm resulting in an overall frequency of 34062 Hz. This
frequency will be displayed as Ref. Frequency.
Signal Nr. 1 is measured for frequency offset. Signal 2 is measured to correct for the eddy
currents generated by the slice-select gradient (measurement of long delays).
The data for a given delay time will be rejected if the amplitude "Ampl" is less than a given
threshold-value (e.g. in case of a bad shim). "Points" in the last column means the number
of ADC-sampling points for the given signal. The shorter the delay time after the gradient,
the shorter must be the total sampling time and hence the number of sampling points.
After the sequence was run for the two slice positions + 50 mm and - 50 mm, the data first
will be added up and then subtracted in order to separate the 0th- and 1st-order term.
Results 0
Gradient Compensation (1st order term)
Fig. 21: ECC compensation 1st order terms

In this example the overshoot tells us that the initial gradient amplitude has to be 0.29 %
higher than the normal value without eddy current compensation. For the Fit Quality the
RMS-value is given.

B0 Compensation (asymmetry)
Fig. 22: ECC compensation 0th order terms

Fig. 23: ECC Measurement Results and Specifications

The first plot of the uncompensated eddy fields shows a smooth curve. The next measure-
ment will establish already a large improvement.
The final iteration, usually after 2 or 3 measurements, shows an oscillating pattern indicat-
ing the eddy-fields have become smaller than the measurement system is able to resolve.

Gradient Delay 3.5
The gradient amplifiers have a delayed response time (tnom to tact), due to mainly the coil
inductance and slightly different for each of the three gradient axes. The Gradient Delay
measurement determines the delay time for each gradient axis. A precise timing of gradient
pulses and RF is very important for good image quality.
Measurement 0
The measurement uses a Spin-Echo-sequence without phase encoding, hence only one
line is measured. The gradient to be tested acts as the readout gradient and is switched on
once between the 90- and the 180-pulse and then again during the readout time of the Spin
Echo.
The first gradient pulse causes a dephase of the magnetization vector in the direction of
the gradient axis. The amount of dephase is proportional to the time-amplitude integral of
the gradient pulse. Subsequently, a 180 refocusing pulse is applied, normally resulting in
a Spin Echo at TE/2 time later. However, due to the dephasing of the vector by the first gra-
dient pulse, the echo will only land in the center of the ADC window (which is also centered
exactly around the TE time point) when an equal rephasing gradient has been applied at
the proper time so that the rephasing is finished at the center of the ADC cycle.
For a non-corrected gradient delay, the two gradient pulses are time-delayed. This does not
influence the dephasing effect of the first gradient pulse, however the delayed second gra-
dient pulse causes a rephasing delay and thus a time-delayed echo signal. From the echo
signal delay and the gradient amplitude, the system can calculate the required correction
time.
Difference Deviation Check

Each gradient is measured at three different gradient strengths - 1.5, 4 and 8 mT/m - giving
three gradient delays for each gradient. For an ideal gradient system, all delays should be
the same.
If the Max. Diff. Deviation tolerance is exceeded, some kind of nonlinearity in the gradient
system might be present.
Results
The new delays and maximum deviation values are displayed in two separate tables. If the
procedure was successful, the values will be saved.
Expert Mode
No extra functionality is offered in the Expert Mode.

Fig. 24: Gradient Delay Measurement and Results
Gradient Sensitivity 3.6
The purpose of this measurement is to measure and establish the correct amplitude scal-
ing (sensitivity) for the actual gradient current in the three axes. When the Gradient Sensi-
tivity value is known, accurate gradient field strengths can be calculated by the
measurement control and the correct gradient amplitude data can be sent to the gradient
amplifier assuring accurate FoV (image size).
Nominal Sensitivity 0
The nominal sensitivity is simply the maximum gradient strength of the Gradient System
(amplifier and coil) divided by the digital resolution of the DAC. In the following example
(Fig. 25 / p. 55), the gradient system has a maximum 40 mT/m field strength and an 20-bit
DAC, whereby 19 bits is the dynamic range since 1 bit is used for polarity (it is a 10V DAC).
This gives a nominal sensitivity of 0.07629 T/m for the least significant bit (LSB) which can
be set. The first Gradient Sensitivity measurement is based on this value.
Actual Sensitivity 0
The actual sensitivity, however, will depend on the characteristics of the DAC, the amplifier
and the coil. In order to determine the actual Gradient Sensitivity, an object of known
diameter (the large 24 cm spherical phantom) is measured using a 500 mm FoV. The
resultant image is evaluated and the actual phantom diameter determined. The actual sen-
sitivity is the old value multiplied by the correction factor which is determined by dividing
the nominal phantom diameter with the actual measured image diameter.
Table Distance 0
During this procedure, the patient table distance - the distance from the light localizer to the
iso-center - is also measured (again, it was first measured some steps ago) and, if neces-
sary, automatically corrected. If a correction is necessary the measurement is repeated.
Measurement 0
The gradsens sequence is a standard Gradient Echo (FLASH) sequence with a TR =

50 msec, TE = 10 msec. The sequence is run three times, one for each orientation. For the
sagittal orientation the usual orientation of the readout and phase encoding gradients are
swapped.
Evaluation 0
The 24 cm phantom is measured and checked initially for size and center.

Results 0
Afterwards, the three values for the Gradient Sensitivity will be stored in the status file. In
addition, the Grad Sens program is used to determine the distance from the light marker to
the Gradient Coil center.
Expert Mode 0
In Expert Mode, the Grad Sens procedure operates similar except that the results are not
written into the database.

Fig. 25: Gradient Sensitivity Measurements and Results
56 Receive Path-Related Tune-Up
4-
General
4Receive Path-Related Tune-Up
The image quality of an MR system is distinguished by its Signal-to-Noise ratio and the
uniformity of the image brightness which are parameters influenced mainly by the quality
of the RF receive coils (coil sensitivity and B1-field homogeneity) and the receive path com-
ponents (amplifier noise figures, etc.).
In addition, each receive coil also will have a specific signal intensity (= mean value of
image brightness) which is determined by the size, gain and sensitivity of the coil. It is, how-
ever, desirable to normalize the signal intensity of the coils so that the image intensity of
the final image is the same regardless which coil is used. We use the pixel value 2000 for
water, the basis of MR imaging.
The signal intensity for any given coil must, therefore, be determined by a software correc-
tion factor RefFFTscale, which is stored in a coil dependant file. This factor is preset for
each coil and is not determined by a Tune-Up procedure.
An advantage of the Tim technology is the ability to combine coils for extended FoVs and
the positioning flexibility of the movable coils (Body Matrix, PA Matrix, Flex coils, etc.). We
have a situation where any coil signal coming off the Patient Table can be directed via the
RCCS switch matrix to any receive channel and connected to different ADCs. This would
result in varying image brightness of the coil depending on the receive channel which is
used. To avoid irregular coil image brightnesses all receive paths must be measured and a
corresponding software scale factors must be calculated - the task off the so called Receive
Path Calibration.
These factors are later used during image reconstruction to correct the image brightness
accordingly.

Receive Path-Related Tune-Up 57
Fig. 26: Receive Path Calibration Measurement and Results
Relative Receive Path Calibration 4.1
Measurement 0
The Relative Receive Path Calibration contains all loop measurements which can per-
formed without MR measurements. The TTX-signal from the Modulator is switched via the
tuning MUX in the RCCS one after the other to every input SGA channel and output chan-
nel of the RCCS. For these measurements the service sequence rx_cal is used - by eval-
uation of raw data amplitudes the software can calibrate the image brightness scale factors
for all receive channel combinations.
In addition to the brightness correction, another factor is calculated for each RCCS input
channel: the relation of signals for high and low receiver gain. This factor (Receiver Gain
Calibration Factor) is needed for sequences where dynamic gain switching of the RCCS
SGAs is used during the measurement. A correction is necessary to avoid image artifacts.
Calibration of RCCS input-matrix

The following steps are performed:
Raw data evaluation from the reference measurement, low/high gain noise, low/high
gain amplitude (graphic representation)
Raw data evaluation from input channel #1, low/high gain noise, low/high gain ampli-
tude (graphic representation)
Measurement data from RCCS input channels 1 - 65
Calibration of RCCS output-matrix

Raw data evaluation from the reference measurement, low/high gain noise, low/high
gain amplitude (graphic representation)
Raw data evaluation from output channel #1, low/high gain noise, low/high gain ampli-
tude (graphic representation)
Measurement data from all available output channels (1 - 32 max.)

Fig. 27: Relative Receive Path Calibration
Results
Fig. 28: Relative Receive Path Calibration Results

The following factors are calculated:
Combined RX Channel Correction - the gain of all available output channels (1 - 32
max.) into the receivers is described
MUX Channel Correction - describes the influence of the different RCCS input chan-
nels (1 - 65)
MuxHighLowGain Calibration Factor - signal amplitude- and phase-difference of the
preamplifiers when they are switched from low- to high-gain (1 - 65)

Absolute Receive Path Calibration 4.2
Measurement 0
The Absolute Receive Path Calibration of the receiver matrix for real MR-signals is deter-
mined by evaluating the brightness of a test image, measured with the Head Matrix Coil.
All previously measured Combined RX Channel Correction factors (1 -32 max.) are
adapted to achieve a reference image brightness value of 2000.
Results 0

Acquisition of an MR-Image with protocol Headmatrix_Tra_sn
Image brightness check against specifications
Combined RX Channel Correction - fine adjustment of the previously measured (Rel-
ative Receive Path Calibration) output channel correction factors (1 - 32max.).

Fig. 29: Absolute Receive Path Calibration
General Quality Assurance 63
5-
5General Quality Assurance
The General Quality Assurance measurements mainly verify the previously adjusted
Tune_Up parameters. In addition, a set of general image quality- and system stability-eval-
uations are performed with the Body Coil.
NOTE The following Quality Assurance descriptions are not com-

plete or in chronological order. They may vary with different
software versions.

64 General Quality Assurance
Image Orientation 5.1
The 6 electrical connections of the Gradient Coil are checked here. If any of the amplifier
or coil connections are swapped or the polarities of any axis is wrong, the images will be
displayed with incorrect orientation - you can imagine the fatal consequences for patient
imaging! Therefore, the Image Orientation procedure has to be performed during Tune-Up
and after repair.
Measurement 0
Two images are generated. The first is a transversal image, the second is tilted from trans-
versal to sagittal orientation by 45 (see next figure).
Evaluation 0
The two measured images are automatically evaluated. If the phantom is not in the
expected position the program eventually makes further measurements to identify the
cause.
In case of incorrect gradient axis orientation, polarity or positioning of the phantom, the pro-
cedure will finish with an "error".
Axis Swap (1st image)

Because the image of the loader is not symmetrical with respect to interchanging of gradi-
ent axes, an exchange of gradient axes (i.e. wrong cable connections) can be detected.
The program searches for the outer edges of the loader from all sides and compares it to
the expected geometry. If correct, it will continue on to the polarity swap measurement. If,
however, a swapped axis is identified, the program terminates with the report of the faulty
axis.
Polarity Swap (2nd image)

Incorrect gradient polarity can be detected by the position of the unsymmetrical fill-cap of
the spherical phantom. If the fill-cap is not in the expected position, the program will make
further measurements with reversed gradient polarities to locate the cap. Then the program
terminates and reports which gradient axes have been swapped. Always remember: there
must be phantom fluid in the fill-cap area, otherwise the measurement will fail!
Expert Mode
No extra functionality is available in Expert mode.

Fig. 30: Image Orientation
Gradient Rise Time Check 5.2
The Gradient Rise Time Check measures the minimal gradient rise time or the so-called
Slew Rate.
The measurement uses a spin-echo EPI sequence; the tested gradient channel has
phase-encoding as well as readout functionality. In the readout interval five bipolar triangu-
lar gradient pulses are switched to generate 5 echoes. The echoes should appear in a nom-
inal position which is defined by the proper pre-dephasing and rephasing of the
grad_risetime sequence. During the measurement, the pulse amplitudes step from -Gmax
to +Gmax.
If there is a deviation of the echo position from the nominal position, the required slew rate
has not been reached.
Measurement 0
The following measurements are performed:

Check if the correct phantom is used
Test sequence grad_risetime is measured using all three gradient axes
Evaluation of the raw data image and calculation of gradient rise time
Results 0
The following results are displayed:

Graphics of echo maximum versus gradient slew rate, echo 1-5
Statistics of measured values against specifications
Y/Z/X- Gradient Rise Time Results
Y/Z/X Gradient Rise Time Multiple Echo Output (raw data image)

Calculation Artefacts 5.3
Fig. 31: Calculation Artefacts

The Calculation Artefacts (CalcAr) QA procedure quantifies the artefact magnitudes (i.e.
signal intensities found outside the image). Causes can be a variety of problems, e.g. gra-
dient deficiencies (instability, non-linearity, 50/60 Hz ripple) or RF instabilities - periodic
blurring or ghosting in the direction of the phase encoding is the consequence.
Measurement 0
A double-echo Spin Echo sequence is run in all three slice orientations (TRA, SAG, COR)
at the iso-center (slice shift = 0). For SAG orientation, the Z-gradient is phase encoding.
Expert Mode 0
In Expert Mode the slice shift 0, 50 mm and single slice orientations can be selected. The
Evaluate option allows to evaluate any image in the actual segment for artefacts.

Spike Check 5.4
The Spike Check is a quick test to check the MR imaging system for the occurrence of
"spikes" and RF interference. The raw data is evaluated by a program that detects and
identifies spikes. The image data is evaluated for RF interference.
Physical Background 0
Electrostatic discharges during measurement can generate "spikes" within the raw data.
"Spike" means individual raw data points or small groups of raw data points with amplitudes
above the typical noise level of the MR imaging system. Spikes can degrade the MR imag-
ing quality. In this case, the signal-to-noise ratio can be reduced or the image can show a
periodic intensity modulation or structure.
One typical source of spikes is the mechanical and electrostatic stress of the MR system
caused by the operation of the gradient system.A simple way to check the MR system for
spikes is to run a noise measurement. This is a measurement that will not generate an MR
signal, but instead pure statistical noise. The spike sequence generates mechanical and
electrostatic stress by switching the gradient system rapidly within the raw data acquisition
period with maximum gradient strength as well as running through tables covering a wide
range of different gradient amplitudes. This sequence applies no RF-pulses and therefore
generates no MR signal.
Expert Mode 0
The Spike Check in Expert Mode is similar except that no results are written into the data-
base. An additional Sequence Advanced Spike is offered for troubleshooting spike prob-
lems that can not be reproduced with the normal Spike Check. Typical examples are spike
problems which may occur only with TRUFI-, HASTE- and EPI-sequences (usage: BOLD-
or Diffusion Weighted Imaging). The "Advanced Spike" should be repeated at least three
times - then the system should be spike-free.
Results 0
The measured raw data and RF-interference images will be loaded and displayed. Image
windowing is done by clicking and dragging the mouse or by direct input of image window
and center values. The same algorithm for spike searching is applied to both image data
and raw data. In case of image data (RF-interference in the image), the spikes are RF-inter-
ference patterns that have been picked up during the noise measurement and then fourier
transformed resulting in a spike in the image data, i.e. the RF interferences will result in
pixels of increased intensity in the image data.
Clicking Log-File in the HTML output page lets you open additional windows displaying the
Spike Check results:
Total number of spike positions
Number of spikes found in real and imaginary part
Number of spikes correlated in real and imaginary part
Position (column, row) of spikes
Amplitudes of spikes in real and imaginary part

Magnitude of the highest spike

Threshold for spike discrimination

Stability Check 5.5
Fig. 32: Stability Check Raw Data

The Stability Check evaluates the stability of the MR signal under various sequence condi-
tions. A Spin Echo and a Gradient Echo are measured without phase encoding using
fixed sequence parameters. Therefore, all raw data lines should look alike. The stability of
the echo signal over the measured lines is evaluated. In standard mode, all three orienta-
tions are measured so that each gradient is performed as a read-out gradient. In addition,
both 0 and +50 mm slice shift positions are measured.

Evaluation 0
Fig. 33: Stability Evaluation

The Spin Echo sequence is sensitive to instabilities in the range of a few Hz up to 100 Hz
which may be caused by 50/60 Hz line hum.
The Gradient Echo/flash sequence is sensitive from DC up to a few Hz which may be
caused by slow external field interferences, field instabilities or mechanical vibrations.
The measurement result of the center slice is sensitive to instabilities in the B0 Field,
whereas the 50 mm offset slice is more sensitive to gradient instabilities.
The amplitude and phase value but also the column of maximum signal (peak column) is
displayed in the graphical output window.

Amplitude and Phase Stability
Fig. 34: Stability Fluctuations

For the Spin Echo, the phase shift is determined by the integral over the magnetic field
from 90 to 180 pulse minus the integral from 180 pulse to echo: For a constant field or a
slowly varying field, both integrals cancel each other, so the measurement is not sensitive
to slow drifts. However, for an oscillating field with a period of TE (20 ms = 1/50 Hz, for
example interference of AC-power lines) this expression becomes maximum - that means
the measurement is very sensitive to AC-fluctuations.
For the Gradient Echo signal, the phase shift is determined by the integral between the
excitation pulse and where the signal is readout. In this case, DC interference and slow
varying fields will contribute to the phase deviation and can be detected. The phase stability
is further evaluated by fitting a straight line to the phase data. Subtracting the value of the
linear fit from the phase data results in the phase stability changes. The slope of the straight
line indicates the linear phase drift.
Expert Mode 0
Expert Mode lets the user select either a Spin Echo or Gradient Echo sequence, modify
the default parameters (slice shift of 0 and +50 mm, orientation, TR and TE values. If the
user selects an invalid TR or TE value an error message is generated and the valid input
ranges will be displayed.

Fat Saturation 5.6
Fat saturation is used during MR imaging to suppress the fat signal while leaving the water
signal unchanged.
Since fat has an MR frequency 3.3 ppm lower than water, this is achieved by applying a low
bandwidth RF-pulse to the fat frequency followed by spoiling gradients.
This requires good homogeneity in the imaging volume in the order of 1-2 ppm peak-peak
to prevent that water is saturated as well. In addition, the RF-flip angle of the Fat Sat pulse
should be about 90 degrees to get optimal results.
Measurement 0
One requirement is to run the test with the standard water filled body phantom. Since no
fat signal is present, no real fat saturation can be tested. However, physically it makes no
difference to saturate water instead of fat, only the frequency of the RF-pulse must be that
of water.
First, a reference image is measured without any saturation. This allows to determine the
100% signal level.
This is followed by a second measurement with water saturation. The quality of fat satura-
tion is now determined by the signal of the saturation measurement divided by the signal
of the normal measurement, which should be as low as possible.
Evaluation of Measured Images 0
First, the normal image which shows the contours of the phantom is analyzed. Position and
diameters are calculated.
Then, a region of interest which is a circle of for example 80 % of the phantom diameter is
taken.
This region is now scanned into the normal image and into the water saturated image, and
the ratio of both signals is calculated point by point.
To reduce noise at every point, a 3x3 average over neighbor points is done. Point by point
saturation ratios are calculated, resulting in a mean and a maximum value.
These two values are compared to specification.
Expert Mode 0
The user can select the three orientations (TRA, SAG and COR) using this mode.
Results 0
The images measured are loaded and displayed. Image windowing is done by clicking and
dragging the mouse or by direct input of image window and center values.
The degree of Fat Sat mainly depends on the shim condition. As a rule, for a mean satura-
tion ratio of 15 %, the peak-peak homogeneity of the field has to better than 4 ppm.

SAR Monitor Test 5.7
The purpose of this test is to verify that the SAR (Safe Absorption Ratio) check is activated.
It is also tested if the pickup-coils detect a signal.
Measurement 0
A measurement with the service sequence gradsens is performed to ensure that the phan-
tom (with loader) has been inserted and centered correctly. The SAR monitor test does not
obtain a quantitative calibration of the SAR but tests that the SAR is switched on. However,
the values retrieved will be displayed.
SW reset of previous values for pick-up coils and SAR value
Measurement with the gradsens-sequence, evaluation of phantom-position and -type
Analysis of SAR and the two pickup-coil values
Display of measured values and the ratio of the two pickup-coil values

Synthesizer Check 5.8
The Synthesizer Check procedure is an important test to verify the proper operation of the
synthesizer. It is used to validate the frequency and phase quality of the synthesizer with
respect to the accuracy of the slice position.
Principle of Synthesizer Test 0
The Avanto system contains a maximum of 32 receiver channels. Four receiver channels
are connected to 1 double (Numeric Controlled Oscillator) NCO, so that we have in sum 16
RX-NCO's. Double NCO means that we have two groups of NCO's, so that it is possible to
switch the group of 4 receiver channels either to NCO-group 1 or to NCO-group 2. For
transmit we have one double NCO. As well the transmit path can be switched to NCO-group
1 or NCO-group 2.
During simultaneous sending and receiving, the phase and frequency of the TX-NCO and
RX-NCO within one group is the same. In the following QA-procedure asynchronous
behavior of the TX- and RX-NCO is needed, that means one NCO is stepping phase and
frequency while the other is fixed. For that reason we always are using different groups of
NCO's when sending and receiving simultaneously, i.e. TX NCO is fixed and RX-NCO is
stepping.
In a first step the verification of the frequency and phase operation of the RF reference sig-
nal is performed by using the MR signal itself (NCO-sequence). After the demodulation the
received MR signal - the reference signal - is compared to the MR signal in terms of fre-
quency and phase. In one test, the reference signal is varied in frequency over the whole
range of the MR system. A second test modifies the phase in fixed intervals. In this way the
whole operating range of the oscillator can be tested.
The test is performed for one RX-NCO from group 1 and one RX-NCO from group 2. So
the two RX-NCO's are validated and can be used for further tests of the other NCO's. In a
second step we perform loop-measurements (nco_loop sequence) between the just tested
RX-NCO's and the TX-NCO's. By doing so, we approve the correct function of the
TX-NCO's. In the further steps all other RX-NCO's are tested against the approved
TX-NCO's by testing separately the group 1 RX-NCO's and group 2 RX-NCO's.
Expert Mode 0
The only difference of Expert Mode is that no data is written into the database.

Temperature Sensor Test 5.9
During every MR examination the Specific Absorption Rate of the body tissue has to be
monitored carefully and checked against the valid country-specific SAR guidelines. In addi-
tion it is important that the temperature inside the bore of the MR scanner lies inside spec-
ified ranges to allow an efficient cooling of the patient. The Air Temperature Measurement
Unit (ATMU) of the Patient Table is responsible for that task and can be tested with the Tem-
perature Sensor Test.
Measurement 0
The procedure works in the following way:

Measurement of initial temperature and check if it is within reasonable limits
A sensor heating element is activated until either a 5 degree temperature rise is mea-
sured or 30 seconds have passed.
Output a pass/fail message
Expert Mode 0
There is no Expert Mode.

Stability_LongTerm Check 5.10
A new application of MR-Imaging is to make activities in the brain visible. For that purpose
the patient will be exposed to a periodical optical, acoustical or other stimulation. During
the exposition, a series of MR-images will be measured. Afterwards the images will be eval-
uated for changes of the signal intensity. The regions of the brain activated by the stimula-
tion can be localized by a very small variation of signal intensity. These variations can easily
be hidden by various instabilities of the system and/or the environment. Therefore, the
instability of the MR-system and interferences from environment must be below a specified
level.
The Stability_LongTerm Check determines the sum of all instabilities while doing a phan-
tom measurement.
Typical causes of instabilities are:
Local B0 field variation caused by temperature variation of the shim plates generate fre-
quency drifts. The result is a virtual movement of the measured object during the scans.
B1 field variations by instable RF produce image intensity variations.
Gradient fields generate frequency and phase variations, causing also virtual move-
ment of the object.
Mechanical vibrations, e.g. by cold head, produce signal intensity variation by the
movement of the measured object relatively to the magnetic iso-center.
Movement of phantom fluid cause phase errors.
Measurement 0
The measurement is done with the Head Matrix coil and the 1.9 l cylinder phantom. A
series of 512 measurements (EPI_FID sequence) with 11 slices will be performed and
repeated once again.
The first measurements is done with a flip angle of 30 degree (evaluation of stability of B1
field, i.e. instabilities in the Tx- chain), the second with a flip angle of 70 degree (minimiza-
tion of unstable RF, detection of B0 and gradient like interferences).
Evaluation of Measured Images 0
The following graphical and numerical evaluations will be performed:

Mean values of a 15x15 ROI related to the mean value of all measurements as function
of image number
Standard Deviation of a 15x15 ROI mean values (linear and logarithmic plot)
Center position in x and y direction of phantom, dependent of image number (Expert
Mode only)
Measurement protocol values

Statistics of Mean values (for ROI size 15x15 and 25x25):

- Mean Image Intensity
- Variation peak-peak
- relative Variation peak-peak
- Standard Deviation
- total relative linear Drift
- Standard Deviation from linear Fit
Image quality parameters such as SNR and Ghosting
Statistics of Phantom Movement (Expert Mode only)

Field Stability Check 5.11
Principle and evaluation of MR-measurement 0
The phase of an MR-FID (Free Induction Decay) is given by:
Fig. 35:
is the gyromagnetic ratio of the proton, B(t) is the time dependent magnetic field. To
replace the integral, it is assumed that the field B(t) does not change much during sampling
of the FID. Time 0 is given by the RF-pulse. So the field can be calculated directly from the
phase at time t. However, because of hardware imperfections it is better to take the slope
of the phase over time which also gives the field.
To measure the field stability over a certain time, a series of FID-measurements is done.
From every FID the phase over time is calculated and a linear fit is done, the slope giving
the frequency.
The time range of the fit should be long to get a high accuracy but is usually limited by the
length of the FID. A good compromise is 20 ms, which averages over 50 Hz line hum which
is usually of no interest in this measurement.
The total measurement time is given by the repetition time and number of FIDs.
Evaluation of Field Stability 0
The result of the field measurement is the field (=frequency) over the measurement time.
As a measure of quality the linear drift over 10 min (=typical measurement time of an imag-
ing sequence) and the maximum deviation from this linear drift are taken. When the mea-
surement time is different from 10 min, the drift is corrected for 10 min by multiplying with
the ratio (10 min / actual measurement time). The drift over 10 min should not exceed the
pixel bandwidth of low bandwidth sequence (e.g. 20 Hz). The second one is important for
Gradient Echo sequences to avoid smearing. To be able to use long echo times it should
be typically less than 2 Hz.

80 Coil Quality Assurance
6-
6Coil Quality Assurance
The QA procedure Coil Check replaces the former Coil QA for all coils except the one
which is used for the Receive Path Calibration in Tune-Up. During this measurement the
Signal to Noise and Image Brightness of the individual coil elements is tested by phantom
measurements.

Appendix 81
7-
Shim
7Appendix
The model: Spherical Harmonics 0
The magnetic field B0 (r, , ) can be expressed as a sum of field terms that are described
by functions known as spherical harmonics:
We measure the B0 field with our Array Shim device on the surface of the 50 cm spherical
volume. A physical rule tells:
When the electric current density vanishes in a finite volume, one can calculate the field
inside this finite volume when knowing the field on the surface of the volume. Hence,
we describe the magnetic field inside our imaging volume with a model using the formula
above. After input of our measurement data, the shim program performs a fit in order to
calculate the terms A(l,m) and B(l,m). Afterward it calculates the deviation = model minus
measurement data. In the output of the shim program this deviation is called residuals.
Hence, the residuals describes how well the magnetic field is described by the model. A
big residual means that something may have been wrong during our shim plot, i.e.:
the field was not yet stable or it changed slowly during the plot because of another rea-
son:
the environment of the magnet changed during the plot (for example: the CSE rotating
the shim device wears a heavy ferromagnetic belt...),
the measurement setup is not correct.
The numbers
A(1,m)
and
B(1,m)
are given in ppm and are calculated by the shim evaluation software using the 480 data
points. Terms up to the 23rd order are calculated. In the above formula spherical coordi-
nates are used. The functions P (l,m) cos are polynomials of cosq, where l is a positive
integer and m = 0,... ,l. They are called associated Legendre Polynomials.

82 Appendix
The functions
and
are given in spherical coordinates and in Cartesian coordinates in the following table:
Term Spherical Coordinates Cartesian Coordinates

A(1,0) r/R cos z
A(1,1) r/R cos cos x
B(1,1) r/R sin sin y
A(2,0) (r2/R2)(1.5cos2 - 0.5) z2 - (x2+y2)/2
A(2,1) (r2/R2) sin cos cos xz
A(2,2) (r2/R2) sin2 cos2 (x2-y2)/2
B(2,1) (r2/R2) sin cos sin yz
B(2,2) (r2/R2) sin2 sin2 xy
where R is the radius of the spherical volume where we measured the field. For a 50 cm
shim volume we have R = 25 cm.
When looking at the table, one might ask why spherical coordinates have been introduced.
For example, everybody knows the meaning of x, y and z: these are the linear terms, which
can also be compensated by the gradient offset currents.
However, the functions of higher order, say A(8,m) or A(10,m) etc. will become more and
more complicated in Cartesian coordinates. The graphical presentation of such high order
terms is much easier and more elegant in a spherical coordinate system. An
A(1,0)
has a very simple symmetry: the field generated by such a term does not change when
moving on a circle around the Z axis. We can also say: there is neither a minimum nor a
maximum on the circle, hence the number 0 for m.
The terms
A(1,m)
with m=1 have one maximum and one minimum etc.

Another important aspect is given by the factor

Appendix 83
1
r
------1-
R
in the beginning of the formula for B0(r, , ), see previous page. r is the distance of any
given point from the center of our 50 cm imaging volume with R = 25 cm, the radius of our
measured sphere.
Fig. 36: Spherical Coordinates
Some examples for inhomogeneities 0
(1) We want to calculate the field generated by the term A(2,0)= 5 ppm in the location
(x,y,z) = (+10.0, +10.0, 0.0) - this is a point on a 28.3 cm-diameter-sphere - using the
definition in spherical coordinates (r2/R2)(1.5cos2 - 0.5). We have r = SQRT(102+102)
= 14.14 and as our point is on a circle around the z-axis with z = 0.0 cm, the angle is
90 resp. cos = 0. Hence the field is:

84 Appendix
It is easy to see that this is the field generated on any point on a 28.2 cm circle around
the Z axis with z = 0.0 cm.
The higher the exponent l for a given term, the less important this term becomes for the
central parts of the imaging volume. For example, when having selected a FOV of
300 mm, a 5 ppm A(12,0) term will play no role, but a poorly shimmed A(2,0) may cause
serious problems for fat saturation.
(2) The next picture shows an example of inhomogeneities on the surface of a spheri-
cal volume (magnets homogeneity is within specification).
Fig. 37: Inhomogeneities on the 50cm spherical volume

Appendix 85
Array Shim Device 0
The Array Shim Device is a service and installation tool for magnet shimming and is not
part of the delivery volume of the MR system. It consists of mechanical supports and a
50 cm diameter disk which can be rotated around the Z-Axis. To measure the field, the
device must be positioned in the iso-center of the magnet
The disk itself contains 24 MR probes which are mounted on the outer diameter of the
50 cm disk. An additional probe in the center is used to determine the reference frequency
at start of the shim plot. Each of these TX/RX-probes contain a sample with hydrogen pro-
tons; a simple FID-measurement is enough to determine the MR-frequency at each posi-
tion precisely.
During the shim procedure, the service engineer has to rotate the disk by 360 in 20 steps,
as a result 2420 frequency values are acquired from the surface of the 50 cm homogene-
ity volume. From that data, the shim program can calculate the magnetic field inhomoge-
neities which are to be compensated with passive and active shimming.
Since there is a limited number of receiver channels on a coil pug, a multiplexer on the
Array Shim Device is used to select the signals sequentially and feed it to coil plug no. 1.
The addressing of the multiplexer is done by the RESET and T_R signal (see next figure).
The selected channel can be seen by yellow LEDs at the front-side of the Array Shim
Device.
The power for the multiplexer comes from the RFIS (-30 V over the coil plug) and from the
RCCS preamplifier supply (+10 V). An additional +5 V voltage is generated by a regulator
at the Array Shim Device. The presence of all three voltages is indicated by green LEDs at
the front-side of the Array Shim Device.

86 Appendix
Fig. 38: Array Shim Device Block Diagram
Changes to previous version 87
8-
8Changes to previous version
First Edition. No previous version exists.

88 Changes to previous version


M6-040 850 20 01 02

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

M6-040 850 20 01 02

Hochgeladen von

Copyright:

Verfügbare Formate

ltr

Print No.: M6-040.850.20.01.02 English

MAGNETOM M6-040.850.20.01.02 Page 2 of 88 Siemens, 2008

1 _______ General ________________________________________________________ 6

2 _______ RF-Related Tune-Up_____________________________________________ 14

3 _______ Gradient-Related Tune-Up________________________________________ 30

Siemens, 2008 M6-040.850.20.01.02 Page 3 of 88 MAGNETOM

4 _______ Receive Path-Related Tune-Up ____________________________________ 56

5 _______ General Quality Assurance _______________________________________ 63

MAGNETOM M6-040.850.20.01.02 Page 4 of 88 Siemens, 2008

6 _______ Coil Quality Assurance __________________________________________ 80

7 _______ Appendix _____________________________________________________ 81

8 _______ Changes to previous version _____________________________________ 87

Siemens, 2008 M6-040.850.20.01.02 Page 5 of 88 MAGNETOM

The Tune-Up is a collection of procedures designed to calibrate, compensate or to deter-

MAGNETOM M6-040.850.20.01.02 Page 6 of 88 Siemens, 2008

The Procedures 1.1

The procedures fall into three major parts:

Siemens, 2008 M6-040.850.20.01.02 Page 7 of 88 MAGNETOM

Siemens, 2008 M6-040.850.20.01.02 Page 9 of 88 MAGNETOM

Each of the Tune-Up steps can be performed in two different modes:

NOTE The Tune-Up procedures in Normal mode measure and auto-

MAGNETOM M6-040.850.20.01.02 Page 10 of 88 Siemens, 2008

Procedure Status 1.3

In a fresh installation all procedures will be set to To Do.

Siemens, 2008 M6-040.850.20.01.02 Page 11 of 88 MAGNETOM

MAGNETOM M6-040.850.20.01.02 Page 12 of 88 Siemens, 2008

Quality Assurance Procedures 1.5

Active Coil Menu 0

Siemens, 2008 M6-040.850.20.01.02 Page 13 of 88 MAGNETOM

Fig. 2: Normalization of Tune Circuit

MAGNETOM M6-040.850.20.01.02 Page 14 of 88 Siemens, 2008

In Expert Mode single measurement steps can be selected and measured.

Siemens, 2008 M6-040.850.20.01.02 Page 15 of 88 MAGNETOM

Fig. 3: Results of Normalizations

MAGNETOM M6-040.850.20.01.02 Page 16 of 88 Siemens, 2008

Body Coil Tuning 2.1

Siemens, 2008 M6-040.850.20.01.02 Page 17 of 88 MAGNETOM

NOTE There is a interdependency between all three adjustments!

There is no Expert Mode for this procedure.

MAGNETOM M6-040.850.20.01.02 Page 18 of 88 Siemens, 2008

Fig. 5: Pre-distortion of RF-pulses

MAGNETOM M6-040.850.20.01.02 Page 20 of 88 Siemens, 2008

Effects of amplitude distortions 0

Effects of phase distortions 0

Siemens, 2008 M6-040.850.20.01.02 Page 21 of 88 MAGNETOM

Fig. 6: RF Characteristic Measurement

MAGNETOM M6-040.850.20.01.02 Page 22 of 88 Siemens, 2008

2. RF Calibration: The characteristic measurement will be made using the DICO, so it

Siemens, 2008 M6-040.850.20.01.02 Page 23 of 88 MAGNETOM

Fig. 7: RF Chara Results

Plot of Measured Characteristics

MAGNETOM M6-040.850.20.01.02 Page 24 of 88 Siemens, 2008

Plot of Calculated Characteristics

Plot of Verification Measurement

Siemens, 2008 M6-040.850.20.01.02 Page 25 of 88 MAGNETOM

RF Characteristic LC (Local Coils) 2.3

Same as for RF Characteristic procedure for Body Coil.

MAGNETOM M6-040.850.20.01.02 Page 26 of 88 Siemens, 2008

2. You have an option to modify the "Maximum allowed RF power".

Siemens, 2008 M6-040.850.20.01.02 Page 27 of 88 MAGNETOM

Coil Power Losses (CPL) 2.4

Fig. 8: Coil Power Losses Measurement

1 _ General __________________________________________________ 6

2 _ RF-Related Tune-Up_______________________________________ 14

3 _ Gradient-Related Tune-Up__________________________________ 30

4 _ Receive Path-Related Tune-Up ______________________________ 56

5 _ General Quality Assurance _________________________________ 63

6 _ Coil Quality Assurance ____________________________________ 80

7 _ Appendix _______________________________________________ 81

8 _ Changes to previous version _______________________________ 87